Lead Scoring Data: Your Score Is Fiction Until Your Deal Values Are Real

A logistics company we work with set a clean target. Bring in a fixed dollar amount of marketing-qualified opportunities every month. Not lead count. Not sessions. Revenue. Honestly, it was one of the sharpest marketing goals a client has ever handed me, because it forced every conversation back to money.

Then the monthly review came around, and the number on screen was a quarter of the target. The problem was that nobody in the room could say whether that number was true. The lead-scoring data underneath it had holes. Several deals sat at zero. One deal was logged at a dollar.

You cannot score leads against revenue when the revenue field is empty. That sounds obvious written down. In practice, almost every company I look at is trying to do exactly that.

The number nobody in the room could defend

Here is what the report showed. Twenty three deals came in that month, carrying a combined value well under the goal. The year before, sixteen deals came in worth four times as much.

On paper, a collapse. In reality, one enormous contract landed in the prior year and skewed everything. Meanwhile, several of the current month’s deals were sitting at zero because nobody had priced them yet, and at least two were described by the sales lead as grossly undervalued.

So the reported figure was wrong in both directions at once. Deals that should have counted were invisible. A single outlier made the comparison meaningless. Frankly, the marketing team was being graded against a number that no one could stand behind.

Where lead scoring data breaks first

Everybody wants to talk about the model. What factors to include, how many points each one earns, where the threshold sits.

That conversation is premature. A scoring model is a prediction of revenue, and predictions need history. If half your closed deals carry no dollar value, you have no history. You have a spreadsheet of guesses with a number attached to some rows.

The first failure is never the algorithm. It is the field that was left blank eighteen months ago because the rep was in a hurry and nothing stopped them from moving on.

A zero and a blank are not the same thing

This distinction matters more than it sounds.

A zero means we looked at this opportunity and decided it was worth nothing to us. Perhaps we turned it away because we could not serve the product. That is a real data point, and your model should learn from it.

A blank means nobody looked. It is an absence, not an assessment. When your CRM treats both as zero in a report, and most do by default, your revenue picture quietly deflates.

The team on that account had a workaround I liked. When a deal genuinely could not be priced yet, they logged it at one dollar. Not zero, not a made up estimate. A dollar, because a dollar is obviously wrong, which means it gets caught on the next review instead of blending into the noise. Small habit, real payoff.

The rep should not be the one guessing

Ask most sales teams how a deal gets its value and you will hear some version of a general formula that everyone applies a little differently.

That is where the lead scoring data quietly falls apart. Two reps look at similar opportunities and produce different numbers, so the score you eventually build on top of those numbers is comparing things that were never measured the same way.

The fix is boring and it works. Whoever prices the work owns the number. When a quote goes out, it comes back with a total value attached, and that value lands in the CRM. Finance or pricing feeds the system, not the rep working from memory between calls.

There is a second habit worth stealing. On that same account, the fields were deliberately left empty rather than pre-populated. Pre-filled fields train people to click next without reading. An empty required field forces someone to actually look at the opportunity. Counterintuitive, but it produces better data than any automation would.

One outlier can make a strong month look like a failure

Once your values are reliable, you still need to read them properly.

A single multi-million-dollar contract in one month will make every following month look like a decline. Of course that comparison is useless. You are not measuring performance; you are measuring the presence of one deal.

Look at the median deal value alongside the total. Track the count of qualified opportunities separately from their combined worth. Then when a giant lands, you can see it for what it is, an outlier worth celebrating, not a new baseline that marketing is expected to hit every month.

Fix these four things before you score a single lead

Here is the order I would work in.

1. Decide who owns the deal value. One team, one method, no exceptions. If pricing owns it, pricing feeds it back into the CRM as part of the quote process.
2. Separate blanks from zeros. Give unpriced deals a flag value that is obviously wrong. Then build a report that surfaces every record carrying it.
3. Run a weekly gap report. List every deal past a set stage with no value attached, grouped by owner. Nobody wants their name on that list twice.
4. Backfill one clean quarter. You do not need three years. One quarter of trustworthy closed won and closed lost data gives you enough to start weighting.

That last one is where most teams stall, because backfilling is tedious and nobody’s job. Do it anyway. Everything the rest of your scoring model does depends on it.

The bottom line on lead scoring data

A lead score is a bet about future revenue. Bets need evidence, and evidence lives in fields somebody has to fill in.

So before you argue about whether a demo request is worth ten points or fifteen, go look at your last hundred closed deals and count how many carry a real dollar value. If the answer is under eighty, your scoring project has not started yet. Your data cleanup has.

That is the unglamorous part nobody writes about. It is also the part that decides whether the model you build predicts revenue or produces a number that looks like insight and means nothing.

Want a marketing program where every lead is tied to a real revenue number? Book a strategy call with NVISION and we’ll map your scoring and reporting gaps in real time.

For more straight talk on marketing, business growth, and what actually drives revenue, follow me on LinkedIn. I share what I’m seeing in the trenches every week.